calice whcal testbeam at sps h8

CALICE WHCAL testbeam at SPS H8

Two testbeam periods• 7 days in June: energies up to 50 GeV• 7 days in July: energies up to 300 GeV

In 2010 a run was taken at the PS T9, with E < 10 GeV

18th July 2011 Erik van der Kraaij, CERN LCD 2

Tungsten HCAL prototype

Scintillator tiles 3x3 cm2 (in centre)Read out by SiPM

Main purpose: Validation of Geant4 simulation for hadronic showers in tungsten

While being installed in SPS H8

Current HCAL setup has 38 W layers.Including active material this is ~4.8 λ

Setting up a beam H8 beam energy is linked to H6 beam energy. • Most of the time we ran with a secondary beam of either

+180, -180 or -300 GeV.


In the beam line we can produce a tertiary beam at lower energy by inserting a target:• 400 mm Cu target for hadron

production• 6 mm Pb target for e+- production.

The "hadron" beam is actually a mixed beam consisting of e+, mu+, pi+, p+• Can be stripped of the e+ with an

absorber of 4 mm, 8 mm or 18 mm thick Pb.

• The muons can easily be identified and taken out in the offline analyses.

Particle identification

H8 beam line has two Cherenkov chambers for particle ID.• The information was that they are 100% efficient from 30 mb

onwards. – This unfortunately turned out to be wrong We put the

pressure slightly above the pion threshold, to be efficient for electrons (no electrons in the pi sample).

– Some 10%-30% of pi in the e-+ sample. Dedicated electron run in second week.


Chamber B

Chamber A

Trigger & tracking setup

Two 10x10 cm2 scintillators in coincidence as main trigger


3 delay wire chambers for track reconstruction

One thick 20x20 cm2 for vetoing against multiple incoming particles, to be used in offline analysis

First event display


+180 GeV

- 40 GeV event display


-40 GeV


Beam profile: 20 GeV

Very broad & low intensity beam, we could not get it more focused. Scanning the low energies with tertiary beam was extremely slow.• With H6 agreed that for the last day we got control of the

secondary beam.

H

V

Wire chamber #1

Wire chamber #2

Wire chamber #3

Low energy scan – pion runs

Energy (GeV) # events180 140k

50 260k40 1.72M30 305k25 220k20 235k-10 215k-20 515k-25 240k-30 630k-40 200k-50 500k


To change the secondary energy one needs to call the CCC, who sets the so-called wobbling of the magnets to the primary target.• Set it to -20, -30 and -50 GeV• At such low energies, -10 or

-25 GeV particles can also be found from the primary target.

• With no secondary target the intensity is much higher– Full scan with negative

polarity was done in 30 hours!

• Total #events taken: ~ 5 Million.

Tailcatcher for high energies


• Installation started immediately after the end of first week run.– Installed and commissioned in the 10 days available. – Only the LED system for calibration does not function

(similar to previous years)

Steel/scintillator-strip sandwich calorimeter with 16 layers (~5.5 λ)

Trigger setupTwo 80x80 cm2cm2 scintillators, one before and one after the calorimeter, are also mounted. Unfortunately not efficient at lower part of detector.• Extended with two 30x30 to scan over full surface for muons.


20x20 cm2 with a hole of 8 cm for vetoing, to be used in the offline analysis.

DREAM instrumentation still in beamline


• The DREAM trigger telescope consists of 6 scintillators. Total of ~5 cm

5 cm

Surrounding material in the beam

For the second week all this equipment was taken out, and a beampipe installed to obtain better electron beam.

Event display including tailcatcher

-180 GeV


Scanning higher energies

With the secondary beam at -180 GeV we set up tertiary beams at lower energies. Rates again too low: • Beam rate @ -100 GeV: 300 counts/spill

Unlike first week, there was not the option of setting up wobblings of lower energies.Tested selecting lower energy particles directly from the secondary beam• Beam rate @ -100 GeV: 3200 counts/spill

Yet this is not how it is supposed to be done at the H8 beamline!• Theoretically, for these high energies, the optics are not

able to select particles with different energies immediately from the focal point on the primary target.

As we would never be able to go through our program with 300 counts/spill, we tried it out 18th July 2011 Erik van der Kraaij, CERN LCD 14

Tertiary vs secondary beam


run # fraction ped

fraction muons

fraction pions

tertiary 10% 25% 65%secondary 0% 35% 65%

electronpion

muon

-100 GeV beam

Check of secondary beam quality: contamination and energy

Tertiary vs secondary beam

-100 GeV -50 GeV

Two energy points crosschecked between tertiary and secondary beam• Same energy distribution Even though it is unclear where the particles originate from

(primary target? Inside TAX?), the particles have the correct energy.

1


Check of secondary beam quality: energy

Pion program

T9


With the unconventional secondary beam, took pion runs with -300 GeV < E < -50 GeV. For all samples at least 600k events.

Many of the runs already reconstructed by Angela! Distributions are from results obtained in the HCAL only (no tailcatcher yet), selecting showers started in the first layer.• For high energy, deviation from linearity is probably due to

punch through

Electron program


On Friday an extra vacuum beampipe was installed in the area where LHCb was also in the beam (with a silicon tracker telescope).• Sunday morning we got control again over the wobbling

settings. Set it to -20, -30 and -50 GeV.

• With the 8 mm Cu target created electron beam at -10, -15, -20, -25, -30, -40, GeV. For each took a sample of 200k events.

As expected, saturation of the SiPM visible:

The T3B experiment

T3B is positioned behind the CALICE W-HCAL • Measures time

development in hadronic shower

New trigger system enables1. high rate standalone

acquisition (10kEv/Spill)2. stable oscilloscope

synchronizationQuite some leakage for > 40GeV nice for T3B

18th July 2011 Erik van der Kraaij, CERN LCD

Fraction of Events in which T3B was Hit. All Energies: 10(pink) to 300GeV(dark blue)

T3B took a nice Data Sample

19

The T3B Experiment in CALICE

Online Calibration was a success All SPS T3B Runs are calibrated: SiPM Gain, Decomposited Wfms, Temperature Information... available• 4 sensors of the Temperature Array Board were located within the

TCMT behind layer 2, 8, 12, 16

T3B Sensors

TCMT Sensors

SPS Test Beam quite successful: T3B took a nice Data Sample We are looking forward to

analyze the dataBig thanks to all CALICE WHCAL members


Other experiments in H8 - LHCb


Most of the time LHCb had their silicon tracker telescope in the beamline. • Were very happy with all the runs that they could join in.• Moved out on Friday morning for the installation of the

beampipe for the electron runs.

Conclusion


• Beam operation took some time to learn, but went very well once understood.– Even with unconventional settings for secondary

beam.– Will try to figure out where the particles originate

from.• Detector performed well.

– Tailcatcher installed, commissioned and operational. Except for LED system.

– Hardly any intervention needed. • We completed our FULL program.

– 14.5 M events in pion runs.– 2.8 M events in dedicated muon runs.– 1.2 M events in electron runs.

• Most samples already reconstructed, first results look good.

Thanks to many people!


Backup

TAX

Primary target

Wobbling


Top view of the end deflection of the beam in the wobbling• As the H8 does not have a horizontal

bending magnet, the H8 beam energy is set by the wobbling: Ewobbling.

H8 secondary beam


Collimators C3 & C9 (horizontal slits, moves vertically) are in the focal points of lenses along the same beamline: act as Δp slits.

Side view of the deflection of the beam to the surface.Bending magnets B1,B2,B3 and B4, and quadrupoles must be set to Ewobbling to propagate particles up to the North Hall.

wobbling

H8 tertiary beam


Adding a target creates a new spectrometer at lower energies.• Intensity is however 10x too

low for us. Would never take enough events in 1 week.

H8 spectrometer


What we did: set B1, B2, B3 and B4 and quadrupoles to settings for particles with energies different than Ewobbling.• We do see particles, the question is

where they originate from. (primary target? TAX? Vacuum chambers?)

wobbling

Beam profile: 40 GeV


Very broad beam, we could not get it more focused

Temperatures• Module 14 has one bad

reading, gets replaced by average of two neighboring layers in next reprocessing.

• Spread of ~3 degC is typical for this run period and similar to T9.

• Plateau in the last layers is not completely understood.


The hot weather end June also influenced the temperature in

the North Hall:

#hits/event distributions• In the #hits /event clear peaks can be seen for different

particle types at all energies– This is one example (no normalization applied)


25 GeV25 GeV30 GeV40 GeV

pedestal

muons

electronsPions/protons

Energy distributions

• In the energy distributions, after the calibration chain, the electron and pion energies overlap.– Selecting now the events where the low pressure

cherenkov chamber did not see a signal, and the high pressure chamber did, we get a pion energy distribution.


25 GeV25 GeV30 GeV40 GeV

muons


• T3B Layer is positioned behind the CALICE W-HCAL

June/July SPS Test Beam quite successful:- Quite some leakage for > 40GeV nice

for T3B- Smooth operation of temperature array

(1 PT1000 sensor for each SiPM + 2 area sensors)

CALICE AHCALT3B Layer

CALICE DAQ

Energy [GeV]

Calice-Sync [kEv]

Standalone [kEv]

+20 221 33+25 205 30+30 290 -+40 2051 1014+50 257 -+180 120 3005-300 820 --250 750 --200 800 --180 750 --150 1000 --120 600 --100 600 --80 700 --60 580 --50 880 --40 192 --30 400 --25 210 --20 410 --15 195 --10 270 -18th July 2011 Erik van der Kraaij, CERN LCD 33


• Live Gain Extraction between Spills working.here, OnlineMonitor without pedestal correction:

• Waveform Decomposition:Use averaged 1photon Waveform to decompose full signal to single photons time distribution

works

T3B is in good shape

Unmodified waveformReconstruced waveform#hits / 0.8 ns


T3B data taking


• T3B has been running the entire week. Took data stand-alone, sometimes included in our DAQ.– Realized the 26th that this lowers our #ev/spill rate. – Their integration in the DAQ is thus removed, they will

match their events to ours in offline analysis.

calice whcal testbeam at sps h8

Documents